It will be observed that each of the flights of stairs was perfectly lighted, the lower and upper being open above, and the intermediate flight open from the side. Their existence here will also explain why the intercolumniation was deeper by one-half in front of the cella than in the flanks. But for this difference, the stairs, instead of being 5 ft. 6 in. in width, could barely have been 2 feet wide. The only other apartment for which it is necessary to find a place in the building is the tomb itself. This fortunately is no difficulty, as the excavated stairs at the west end of the building, and the big stone which was found there, certainly indicate its whereabouts, even if they do not actually fix the spot. Besides this, the expressions used by Guichard in themselves almost suffice—“It was situated beyond a low doorway, after the manner of an antechamber.” This cannot, of course, apply to a vault under the hall first discovered by the Knights, but describes accurately such a chamber as the wider intercolumniations at the further end would fully admit of, while the fact of the stairs being excavated[21] gives the requisite height without interfering with the peristele above. In the plan and sections I have suggested stairs leading down to it; and even if it is insisted that the Tomb of Mausolus, on the right, was walled up,[22] and the stones let down immediately after the interment, it does not follow that the Tomb of Artemisia, which probably was on the left, may not have been accessible long afterwards; and there may have been other vaults beneath to which it was desirable to give means of access. There may also have been recesses for sarcophagi or urns in the thickness of the walls on either side of the principal chamber, as represented in the plan; but these are details it is hardly worth while entering into at present. There is no authority for them, so every one may supply or reject them as suits his own fancy. _Lacunaria._ One further merit of the restoration just described is, that it entirely gets over the difficulty of the Lacunaria of the peristyle, which rendered Lieut. Smith’s proposal so inadmissible. With the arrangement of the columns here suggested, and the dimensions obtained for the cella, the greatest width to be spanned in front and rear is only 14 Greek feet—2 feet 8 inches less than Mr. Pullan makes it. Although it is just such an increase as this that makes the difficulty in most cases, neither of these dimensions ought to be considered insuperable, inasmuch as in the Propylæa at Athens a marble roof is thrown over a clear space of 18 feet 6 inches English; and though it may be suggested that the roof over these Lacunaria was lighter, that does not alter the case. No part of the external roof of the Mausoleum rested on these beams, and they therefore were not affected by its weight.